AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Revisiting Feature Disentanglement Strategy in Diffusion Training and Breaking Conditional Independence Assumption in Sampling

Wonwoong Cho, Hareesh Ravi, Midhun Harikumar, Vinh Khuc, Krishna Kumar Singh, Jingwan Lu, David Iseri Inouye, Ajinkya Kale

ECCV 2024 · Image generation › Fast diffusion sampling · approach: Diffusion models as perception tools

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add joint guidance via GCDM (lambda=1.0) instead of independent CDM guidance (ablation)FDiff · AFHQFID: 10.57 -> 8.58Improved
increase joint guidance weight lambda in GCDM (ablation)FDiff · AFHQLPIPS: 0.59 (CDM) -> 0.57 (lambda=1.0)Mixed
use GCDM instead of CDM without timestep scheduling (ablation)FDiff · FFHQFID: 21.43 -> 14.46Improved
add timestep-dependent weight scheduling to CDM (ablation)FDiff with CDM · FFHQFID: 21.43 -> 10.50Improved
add timestep-dependent weight scheduling to GCDM (ablation)FDiff with GCDM · FFHQFID: 14.46 -> 10.21Improved
add timestep scheduling (ablation)FDiff with CDM · FFHQLPIPS: 0.47 -> 0.31Hurt
use GCDM instead of CDM for Stable Diffusion text prompt compositionStable Diffusion V2 · multi-object prompts (qualitative)composition realismImproved
lower lambda to weaken joint guidance (with beta_s=1) (ablation)FDiff · FFHQFID: 11.99 -> 15.45Hurt
lower lambda to weaken joint guidance (with beta_s=1) (ablation)FDiff · FFHQLPIPS: 0.34 -> 0.49Improved
use FDiff with two disentangled features instead of baselines for reference-based translationFDiff · FFHQFID: 25.06 (SAE) -> 11.99Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy